2024-05-22 View 985

Seoul AI Business Pledge(eng)

We, as technology companies, acknowledge the rapid acceleration of technological advancements in AI and its convergence with the broader economy and society, and their growing impact on the global community.


We acknowledge our roles and responsibilities in fostering the AI industry ecosystem and creating new added value by applying novel AI technologies, while supporting social progress and enhancing the quality and convenience of human life by providing AI products and services.


We affirm the importance of the values of Safety, Innovation, and Inclusivity highlighted in the Seoul Declaration adopted at the AI Seoul Summit, recognizing their significance in advancing AI governance.


We support the efforts of each nation in upholding these values to fulfill their commitments.


We recognize the “Code of Conduct” from Hiroshima AI Process, the United Nations General Assembly resolution “Seizing the opportunities of safe, secure and trustworthy artificial intelligence systems for sustainable development”, and the “Frontier AI Safety Commitments” announced through “AI Seoul Summit”.


We will work to ensure responsible AI development in line with these initiatives. We commit to upholding the three strategic priorities, through our efforts including advancing AI safety research, identifying best practices, collaborating across sectors, and helping AI meet society’s greatest challenges.


Ensuring responsible development and use of AI We commit to continuing to advance research endeavors to promote responsible development of AI models, minimize risks, and enable robust evaluations of capabilities and safety throughout the AI system life cycle, where appropriate.


We further commit to drawing on international best practices for safe, secure and trustworthy AI.

We commit to put in place robust internal governance and risk management policies for AI development that (i)identify andevaluate vulnerabilities and risks, (ii)integrate results from internal evaluations into the AI model lifecycle, and consider results from external evaluations such as those by home governmental entities like an AI Safety Institute, where appropriate, and (iii)monitor and respond to AI safety incidents.


We commit to strengthening collboration across AI actors, such as networks of AI Safety Institutes, policymakers, international organizations, private companies, academia and civil society.


This includes enhancing cooperation to develop international standards, sharing best practices for AI safety test frameworks and red-teaming.

We commit to earning and upholding the public's trust by protecting our shared values. In the stages of development, training, and operation of AI, we will endeavor to implement appropriate safeguards to respect applicable domestic and international legal framework.


We will work towards the development of methodologies where appropriate, such as watermarking, to enable users to identify content generated by AI, and to pursue measures to mitigate the use of AI-generated content for dis- and misinformation, which threatens democracy.


Pursuing sustainable development and innovation in AI We commit to sustaining our investment in developing state-of- the-art AI models and systems, which have the potential to
tremendously expand human knowledge, enhance human well-being,and create new economic opportunities.


We strive for our AI innovations to positively transform people’s lives.

We commit to fostering an enabling environment for AI including by promoting research and development, strengthening partnerships, and raising awareness and understanding of the core functions, capabilities, limitations and impact of the technology.


Aiming for the advancement of the entire AI ecosystem, we will collaborate with SMEs and startups. Actions to this end could include, for example, joint projects as well as support programs such as innovation funds and technical consulting.


Additionally, we commit to expanding inter- industry cooperation to integrate AI into traditional industries, thus revolutionizing productivity and creating new value-added

We also commit to fostering the education and cultivation of professional talents. Programs such as internships, specialized educational and workplace support can be counted as examples of activities supporting these goals. Furthermore, we will contribute to the growth of future generations and job creation by establishing collaborative networks between academia and industry, and will alsoexpand the opportunities for an exchange of technological achievements through the networks.


Ensuring the equitable benefits of AI for all We affirm our support for governments’ efforts to ensure equitable access to digital infrastructure including computing resources, and to facilitate dissemination and distribution of high-value data.


In light of this perspective, we commit to joining the international efforts to enhance capacities to develop and utilize AI, and support for resources and infrastructure to research labs and universities, thereby bridging these divides and providing new opportunities for developing countries and emerging economies.


We stress the importance of developing AI services in a manner that leaves no one behind including marginalized regions and communities, in cooperation with governments and global entities across the broader ecosystem. We commit to launching AI services to enhance the convenience and the quality of daily lives of the people especially including socially vulnerable groups.


We also commit to developing and/or collaborating on innovative AI solutions to address societal issues, and help solve global challenges.


We will endeavor to ensure that these solutions embrace diversity, promote inclusivity, and are regionally and culturally appropriate as well as responsive to local needs.




The following is the list of companies agreed to join the Pledge:


· Adobe


· Anthropic


· Cohere


· Google




· Kakao


· KT


· LG AI Research


· Microsoft




· OpenAI


· Salesforce


· Samsung Electronics


· SK Telecom


· Cisco Systems (Joined 11 July 2024)